J. Org. Chem. 1985,50, 3336-3341
3336
added 5 mmol of the sulfone, 3.0 mL (15 mmol) of anisole, and 0.88 mL (7.5 mmol) of triflic acid. The reaction mixture was stirred a t 150 "C for 4 h, allowed to cool to room temperature, and poured over 50 g of ice. The mixture was extracted with chloroform and the chloroform extract was washed sequentially with 40 mL of water, 40 mL of saturated NaHC03, and 40 mL of water and then dried over CaC12. The dry chloroform extract was then subjected to quantitative GC/MS analysis. This procedure was repeated as above except that the reaction mixture was separated by column chromatography using silica gel in a 50 cm X 2 cm column. The product sulfone was eluted using a diethyl ether-chloroform (v:v 29) mixture. Evaporation of the solvents gave methyl p-anisyl sulfone [mp 120-121 OC;12mass spectrum, mle 186 (M'), 171,123, 107,94, 771 and phenyl p-anisyl sulfone [mp 90-91 OC;13mass spectrum, m / e 248 (M'), 177,123,107,771. General Procedure for the Reaction of Sulfones with Diphenyl Ether. The procedure was exactly as described above except that diphenyl ether was used in place of anisole. Methyl p-phenoxyphenyl sulfone [mp 85-86 "C;14 mass spectrum, m / e 248 (M'), 233,185, 169,141, 771 and phenyl p-phenoxyphenyl sulfone [mp 92-93 "C;15 mass spectrum m / e 310 (M'), 233,217, 185, 169,771 were obtained as transsulfonylation products. The identity of these known products was confirmed by NMR as well as by melting point and MS. General Procedure for the Reaction of Methyl Mesityl Sulfone with Various Arenes. The procedure was as described above except that various arenes (15 mmol) were used in reaction with methyl mesityl sulfone (5 mmol) and triflic acid (7.5 mmol).
Analyses by GC/MS and column chromatography were carried out as before. The results are presented in Table I. The product sulfones formed by transsulfonylation in experiments 2 and 3 were the same ones used as starting materials in some of the experiments described above. The fact that the retention time in the capillmy GC of the product of experiment 2, Table I, was the same as that of the pure methyl p-tolyl sulfone used as starting material previously indicated that the product sulfone was indeed the para isomer; there was no indication of more than a trace of other isomers formed in the transsulfonylation. The product sulfones formed by transsulfonylation in experiments 4 and 6, Table I, were the same ones produced in previous experiments. The product sulfone of experiment 5, Table I, ethyl p-ethoxyphenyl sulfone, had mp 89-90 "C,16 mass spectrum, mle 200 (M'), 185, 137, 121, 77. Tests for Desulfonylation in the Absence of a Nucleophilic Arene. Each of the seven aryl sulfones was stirred with triflic acid (7.5 mmol) at 150 "C for 3 h and the reaction mixture was worked up as before. Each of the starting sulfones was recovered in almost quantitative yield by evaporation of the chloroform solution.
Acknowledgment. Support for this work by the Robert A. Welch Foundation is gratefully acknowledged. Registry No. 5, 97416-12-1; 6, 6462-31-3; 7, 3112-82-1; 4MeS02C6H40Me,3517-90-6; 4-PhSO2C6H4OMe,3112-84-3; 4MeS02C6H40Ph, 21134-15-6; 4-PhSO2C6H4OPh,47189-05-9; 4-EtS02C6H40Et,82961-62-4; 4-MeS0,C6H4Me, 3185-99-7; anisole, 100-66-3; diphenyl ether, 101-84-8; toluene, 108-88-3; mxylene, 108-38-3;phenetole, 103-73-1;isodurene, 527-53-7; methyl isoduryl sulfone, 97416-13-2.
(12) Gilbert, E. E. J. Org. Chem. 1963, 28, 1945. (13) Hilbert, J. E.; Johnson, T. B. J. Am. Chem. SOC. 1929,51,1526. (14) Arcoria, A.; Marziana, N.; Passerini, R. Gazz. Chim. Ital. 1961, 91,223. (15) Campbell, J. R.; Hatton J. Org. Chem. 1961, 26, 2480.
(16)
Suter, C. M.; Hansen, H. N. J. Am. Chem. SOC.1932, 54, 4100.
Catalysis of Sulfonate Ester Hydrolysis by Intramolecular Amide Group Assistance Sergio Thea,*t Giuseppe Guanti,? Andrew R. Hopkins,* and Andrew Williams** Istituto di Chimica Organica dell'Uniuersit6, C.N.R., Centro di Studio sui Diariloidi e loro Applicazioni, Corso Europa, 16132 Genoua, Italy, and University Chemical Laboratory, Canterbury, England Received August 3, 1985
Kinetics of the alkaline hydrolysis of aryl 2-(acylamino)benzenesulfonates (1, X = OAr) obey the equation k,b,.j*(k,
t kb[OH-I
)/(
I t [H'I/K,)
Hammett equations correlate the parameters k,, kb, and K , for variation in both amido and leaving phenolate substituents. The values and sign of the p values together with entropy of activation data, reactivity, trapping, and oxygen-18 incorporation are consistent with the formation of an intermediate benzoxathiazine S,S-dioxide (2). The k , term involves intramolecular attack of the amido anion. The kb term is consistent with a specific anion effect on k,. Regular bimolecular BA,2mechanisms for k , and k b are not consistent with the high observed reactivity of these parameters.
The amide group is a well-known intramolecular nucleophilic catalyst for many ester hydrolyses,' where it can act in its neutral or coniugate base forms. Previous work from these laboratories his indicated that a neighboring oxyanion is a powerful nucleophile in sulfonyl group t r a n ~ f e r . ~ dWe are interested in the amido anion as a potential neighboring group for sulfonyl transfer where the Istituto di Chimica Organica dell'universitl. University Chemical Laboratory. 0022-3263/85/1950-3336$01.50/0
reacting nucleophile is the oxyanion function. Aberlin and Bunton4 showed that the hydrolysis of 2-acetamido(1) (a) Cremin, D. J.; Hegarty, A. F. Tetrahedron 1977, 33,1823. (b) de Jersy, J.; Willasden, P.; Zerner, B. Biochemistry 1969, 8, 1959. (c) Williams, A. J. Chem. Soc., Perkin Trans. 2 1975,947. (d) Hegarty, A. F.; Frost, L. N.; Coy, J. H. J. Org. Chem. 1974, 39,1089. (e) Morawetz, H.; Shafer, J. A. Ibid. 1963, 28, 1899. (2) Deacon, T.; Farrar, C. R.; Sikkel, B. J.; Williams, A. J. Am. Chem.
SOC. 1978, 100, 2525. (3) Farrar, C. R.; Williams, A. J. Am. Chem. SOC.1977, 99, 1912. (4) Aberlin, M. E.; Bunton, C. A. J. Org. Chem. 1970, 35,1825.
0 1985 American Chemical Society
Catalysis of Sulfonate Ester Hydrolysis benzenesulfonyl fluoride (1, R =
1
CH,;X = F) probably
2
involved neighboring group participation by the amido base to form benzoxathiazine-S,S-dioxide(2, R = CH,). Duchek5 synthesized this species and showed that i t was very reactive toward nucleophiles, consistent with its being an intermediate in t h e fluoride hydrolysis. We examine the alkaline hydrolysis of aryl sulfonate esters (1; X = OAr) using oxygen-18 incorporation and structure-reactivity relationships to indicate that the species 2 is a probable intermediate.
Experimental Section Materials. Substrates were prepared by standard chemical techniques according to pathways given in eq 2 and 3 for the 2-(acylaminobenzenesulfonates. Examples of the preparations
are given below for the synthesis of 4-nitrophenyl2-acetamidobenzenesulfonate. Phenyl 2-nitrobenzenesulfonate was prepared by adding a dichloromethane solution of 2-nitrobenzenesulfonyl chloride (3.5 g, 15.8 mmol; prepared by the method of Vogele) to an ice-cold solution of phenol (1.49 g, 15.8 mmol in 20 mL of dichloromethane. The mixture was stirred overnight and then worked up with diluted HCl(4%), and the dichloromethane phase was dried and evaporated. The residue was recrystallized from ethanol and melted at 55-57 "C. Reduction of the nitro group was carried out as follows. The ester (5.6 g, 20 mmol), concentrated HCl (10 mL), ethanol (8.5 mL), and dioxan (12 mL) for solubilization of the phenyl ester were maintained at 60 OC and stirred while granulated tin (4.5 g, 38 mmol) was added. The rate of addition was controlled to maintain the temperature at 60 OC. The mixture was stirred for an hour after all the tin had been added and then poured into an ice-water mixture. A solid separated, which was dissolved on addition of NaOH (2 N), the alkaline solution was extracted with ether, and the ether layer was dried and evaporated. The residue, after evaporation, was recrystallized from aqueous ethanol and had mp 72-37 "C (yield 85%). The 2-amino function was acetylated by standard procedures with acetic anhydride. The crude product was used directly to prepare the 4-nitrophenyl ester. Phenyl 2-acetamidobenzenesulfonate (0.17 g, 0.58 "01) was dissolved in H S 0 4 (3.16 g) and cooled in ice. A mixture of KNOB(0.059 g, 0.58 "01) in HZS04(2.7 g) was added dropwise to the above solution, which was then kept at room temperature overnight with stirring. The mixture was then added to ice and extracted with chloroform, and the dried organic phase was evaporated. The residue, after recrystallization from ethanol, had mp 72-73 "C. 4-Benzamidobenzenesulfonyl chloride was prepared by the method of Bere and Smiles' and used without purification to prepare the corresponding 4-nitrophenyl ester. (5) Duchek, J. R. Ph.D. Thesis, St. Louis University, MO, 1975. (6) Vogel, A. I. "Practical Organic Chemistry";Longmans: London 1961, p 542. (7) Bere, C. M.; Smiles, S. J. Chem. SOC.1924, 2359.
J. Org. Chem., Vol. 50, No. 18, 1985 3337 Phenyl 2-(N-methylacetamido)benzenesulfonatewas prepared by adding methyl iodide (0.36 g, 2.5 mmol) to a boiling solution of phenyl 2-acetamidobenzenesulfonate(0.5 g, 1.7 mmol) containing finely powdered KOH (0.17 g, 3 mmol) in anhydrous acetone (7 mL). A white precipitate formed immediately, the mixture was refluxed a further 5 min, and then the solvent was evaporated. The residue (0.45 g) was washed with water and used directly after drying. TLC analysis indicated the absence of byproducts in the reaction. Standard nitration with KN03/H#04 benzenesulfonate. gave the 4-nitrophenyl2-(N-methylacetamido) 2-Acetamidobenzenesulfonicacid (sodium salt) was prepared from sodium orhanilate by using acetic anhydride. The product was recrystallized from 95% ethanol and had mp ca. 290 "CS Materials used for kinetics either were of analytical reagent grade or were recrystallized or redistilled from bench grade materials. Water used throughout was glass-distilled deionized material. Dioxan was purified by percolation of the AR grade material through activated alumina; the absence of peroxide impurities was checked with KI. l8O-Enriched water was purchased from Prochem Ltd. The substrates were routinely checked for purity by TLC on Merck silica gel precoated plates, generally by using CHZCl2or CHZCIZ/ethylether mixtures for elution, and structures were confirmed by IR and NMR spectroscopy. Satisfactory analyses were obtained for new species (microanalyses were carried out by the Institute of Pharmaceutical Chemistry at Genoa). Analytical and spectroscopic data are reported in Table S1 (supplementary material) for the new species. Methods. 20% Aqueous dioxan (v/v) was used as solvent for kinetic and pK, measurements because of the low solubility of the substrates in water. Values of pK, for this solvent were obtained from l i t e r a t ~ r e . ~Kinetics were measured spectrophotometrically with either a Perkin-Elmer 554 or a Gilford 2400 S instrument coupled with potentiometric recorders and fitted with thermostatically controlled cell holders. A typical experiment involved adding an aliquot of substrate (25 pL) dissolved in dioxan to 2.5 mL of buffer in a silica cell in the cell compartment of the spectrophotometer. The absorbance was then measured against time at a fixed wavelength previously determined from a separate spectral scan of the reaction. The pH of the buffer in the silica cell was measured before and after the reaction with a Radiometer PHM 62 pH meter calibrated with Merck standard buffers to 10.02 pH unit. In some cases the pH of the solution in the cell was maintained constant by the use of a pH-stat apparatus simiiar to the one described previously.1° The pseudo-first-order rate constants were obtained from plots of A , - A, against time using two-cycle semilogarithmic graph paper. Reactions were normally followed to about seven half-lives. Ionization constants were obtained for substrates by measuring the UV spectra as a function of pH and the pK, determined from a plot of pH vs. log FA/FB (eq 4) where FA and FB are the pH = pK, - log FA/FB
(4)
fraction of acid and base species, respectively, as determined from the absorbance measurements at an appropriate wavelength in the spectrum. Product analysis was carried out on the hydrolysis of 4-nitrophenyl 2-benzamidobenzenesulfonate. The ester (0.2 g) dissolved in peroxide-free dioxane (5 mL) was added to an aqueous solution of KOH (0.2 M, 5 mL). The solution was stirred until TLC analysis indicated that the reaction was complete. The mixture was acidified to pH 4 and extracted with ether (3 X 15 mL), and the aqueous phase was evaporated under reduced pressure. The product was treated with the nitrous acid12-naphthol test" and shown to contain no aromatic amine. NMR spectroscopy in D20 indicated 2-benzamidobenzenesulfonicacid (potassium salt) as the only product other than 4-nitrophenol. An identical experiment was carried out with the 2-acetamido ester. In this case 2-acetamidobenzenesulfonicacid potassium salt prepared separately was used as a standard in the TLC analysis. (8) (a) Claasz, M. Justus Liebigs Ann. Chem. 1908, 380, 306. (b) Grammatikakis, P. Bull. SOC.Chim. Fr. 1964, 92,97. (9) Harned, H.S.;Fallon, L. D. J. Am. Chem. SOC.1939, 61,2376. (IO) Thea, S.;Williams, A. J. Chem. SOC.,Perkin Trans. 2 1981, 72. (11) Schetty, G.H e l a Chim. Acta 1966, 49, 461.
3338 J. Org. Chem., Vol. 50, No. 18, 1985
Thea et al.
Table I. Physical and Ionization Constants for the Substrates at 25 'Cf substrate 1
R C6H5 C6H5
C6H5
C6H5
C'H, CiH; C6H5
NC
pKakin
12.26
8
12.35 12.39
330 330 330
12.35 12.35 12.52
8 8 8
330 330 330 330
12.35 12.65 12.65 11.52
12 7 7 8
310 310
13.09 13.02
7 9
320
13.28
9
X
w/Ma
h/nmb
pKasPeCtr
4-N0,C6H,0 4-NO,C6H,O 4-CNC6H,0 3-C1C6H,0 3-BrC, H,O 3-BrC;H;O 3-CNC6H,O C,H.O 4-uN0,C,H, 4-N0,C6H, 4-N0,C6H,0 4-NO,C,H,O 4-N0,C6H,0
0.2 1.0 1.0 0.2 0.2 1.0 1.0 1.0 1.0 1.0 1.0 0.2 1.0
330
C,H. 4kH,C,H, 3-C1C,H4 3-NO,C H, CH3 CH3 other substrate
1.0
qn,con"~soooc.u..4-NOI
mp/"Ce
d
116-116.5 96-97 96-97 115-116
12.37 12.65 12.60
107.5-109 100-101 131-132 124-125 179-181 72-73
12.62 11.62 10.87 12.92 13.01
190-191
Number of points for a Ionic strength made up with KCl. Wavelength for spectroscopic determination of pK,. This value is derived from the kinetic data (for condispectroscopic determination of pK, (not including duplicates). tions, see Table 11). e Solvent used for recrystallization is ethanol except for the 3-chlorobenzamido ester for which methanol was used. f Value of pK, for these conditions is 14.62.9
4-Nitrophenyl2-acetamidobenzenesulfonate was hydrolyzed in 180-enrichedwater to check the position of any labeling in the acetyl function as shown in eq 5. The ester (0.17 g) in dioxan
a
NHCOCH3 CH, ~
N
-
H
0.2
0.4
0.6
I
,
1.5
CH3-@NHMCH3
(5)
SO3-K'
kOHl/ M
(5 mL) was added to aqueous KOH (0.2 N, 5 mL; containing enriched water at 4.64% '80 enrichment). The mixture was stirred at room temperature for 20 min until TLC analysis indicated that hydrolysis was complete. It was then acidified to pH 4 with condentrated HCl and extracted with ether, and the aqueous phase was evaporated under reduced pressure. The white solid (0.5 g), containing mostly KC1 and potassium N-acetylorthanilate (checked by NMR spectroscopy), was added to 4-toluidine (1.13 g), and the mixture was refluxed under argon for 2 h (oil bath at 220 "C (eq 5)). Water was added to the cooled mixture, and then concentrated HCl was added in order to neutralize the excess amine. The solution was finally extracted with ether to give the product N-acetyl-4-toluidine, which contained only traces of impurities as judged from TLC analysis. An analytical sample obtained by recrystallization from ethanol (mp 148-149 "C, lit.l* mp 146 "C) was subjected to mass spectral analysis (AEI MS 902 high-resolution mass spectrograph) under the supervision of Dr. J. F. J. Todd. Trapping with ammonia was carried out as summarized in Table V by allowing 4-nitrophenyl2-benzamidobenzenesulfonate to react in buffers with increasing ammonia concentration. Reaction rates were measured, and the UV spectra were recorded after completion of the reactions and used to analyze for trapped products. The 4-nitrophenolpeak at 400 nm provided an excellent internal check on stoichiometry. It proved difficult to study the trapping reaction by TLC analysis although this was attempted by using benzyl amine as the trapping agent instead of ammonia. Reaction rates were measured at different temperatures for the hydrolysis of the 4-nitrophenyl benzamidobenzenesulfonates under different conditions in order to obtain values of the activation parameters. The relevant rate data are reported in Table S2 (supplementary material).
Results Product analysis under the conditions of t h e hydrolytic kinetics at high pH revealed that only t h e sulfonate ester link is cleaved in t h e case of t h e 4-nitrophenyl esters of (12) Heilbron, I., Cook, A. H., Bunbury, H. M., Hey, D. H., Eds. "Dictionary of the Organic Compounds"; Eyre and Spottinswoode: London, 1965.
0.8
1.0
I
I
-
0.004 0.008 0.012
0.016
092
Figure 1. Dependence on hydroxide ion concentration of the pseudo-first-order rate constant for hydrolysis of 4-nitrophenyl 2-(3-chlorobenzamido)benzenesulfonateat 25 "C. Ionic strength maintained at 1 M with KCl. Points for very low concentrations (a)are on the expanded scale. The lines are calculated from eq 6 with parameters given in Table I and 11. the 2-acetamido and 2-benzamidosulfonic acids. T h e very sensitive diazonium salt/2-naphthol test indicated n o free amino group production. These results contrast with those found with t h e fluoride derivative.* The rate of hydrolysis obeyed excellent pseudo-firstorder kinetics u p t o 90% of t h e total reaction. T h e pseudo-first-order rate constants varied with hydroxide ion concentration according t o e q 6
except when there was n o ionizable NH group. I n t h e latter cases t h e rate constants were overall second order, i.e., first order in both substrate and hydroxide ion concentration. T h e data for the ionization constants (K,) and rate parameters are collected in Tables I a n d 11, a n d a n example of t h e hydroxide ion concentration dependence is illustrated in Figure 1. Some ionization constants (K,) were obtained spectrophotometricdy as well as kinetically, a n d Table I indicates excellent agreement between t h e results of t h e two methods. A typical plot of absorbance vs. p H used for t h e spectrophotometric determination of pK, is shown in Figure 2. T h e values of k,, k b , a n d K,
J. Org. Chem., Vol. 50, No. 18, 1985 3339
Catalysis of Sulfonate Ester Hydrolysis
Table 11. Rate Parameters for the Hydrolysis of the Substrates a t 25 "C substrate 1
X
dMb
nm
NC
[OH-] range, M
104k2/
C6H5
4-NO2C,H,O 4-N0,C,H40 4-CNC,H40 3-C1C6H,0 3-BrC,H40 3-BrC,H40 3-CNC,H,O
C6H5
C6H50
0.2 1.0 1.0 0.2 0.2 1.0 1.0 1.0 1.0 1.0 1.0 0.2 1.0
400 400 275 275 275 275 248 320 400 400 400 400 400
8 10 11 8 6 8 10 4 16 14 15 6 10
0.005-0.1 0.005-1.0 0.01-0.2 0.001-0.2 0.005-0.2 0.005-1.0 0.005-1.0 0.005-1.0 0.005-1.0 0.001-1.0 0.00012-1.0 0.005-0.2 0.005-1.0
300 364 67.5 1.52 1.97 2.15 12.1 0.114 537 103 36.6 24 5 293
a
1.0
400
6
0.02-1.0
C & C O N H ~ S O z O € ~ H . - 4 -NO2
1.0
400
6
0.02-0.75
2.8
C6H~SOzOC6H,-4-N0,
0.2
400
4
0.02-0.2
2.6
R C6H5 C6H5 C6H5 C6H5
C6H5 C6H5
4-CH,C,H4 4-NO,C6H,O 3-C1C,H4 4-NO,C6H,O 3-NOzC6H, 4-N0,C6H40 4-N0,C,H40 CH, I-NO,C,H,O CH, other substrates
S-'
104kg/ S-l
119 22.5 0.46 5.36 0.021 166 43.5 13.8 127
1O2(k&K,/K, M-I s - l
I/
68 7 618 126 2.25 22.5 0.106 501 1300 2060 117
NICH3)COCHa'
2.85
S q w - 4 - N O g
These rate constants are unambiguously bimolecular. Ionic strength made up with KC1. including duplicates. Errors o n these values are not greater than 5%. e Mp 123.5-124.5 "C. K. T.; Loran, J. S.; Steltner, A.; Williams, A. J . A m . Chem. SOC.1977,99, 1196. a
7.2
0.613
3.5
1.6°7f
Number of points not f Davy, M. B.; Douglas,
Table 111. Correlation of Kinetic and Equilibrium Parameters with Hammett's u and uNHCOPh
0.2
S020Ar
d
log k, = (2.86 f 0.18)~-- (4.81 f 0.14) log k b = (3.08 f 0.30)~-- (5.49 f 0.22) pK,b = (-0.46 f 0.11)~+ (12.68 f 0.06)
r = 0.994c r = 0.986' r = 0.809
NHCOAr
SOgOCsH4.4- NO2
log k, = -(1.35 f 0.05)~- (1.48 f 0.02) log k b = -(1.24 f 0.08)~- (1.95 0.03) pK, = 42.02 f 0.12)~+ (12.31 f 0.02)
*
-3
-2
-1
LOG [KOH]/M
Figure 2. The pH dependence of the absorption a t 330 nm of 3-cyanophenyl 2-benzamidobenzenesulfonate a t 25 'C. Ionic strength maintained at 1M with KC1. The line is calculated from the pK, parameters in Table I.
fitted Hammett relationships satisfactorily. The relationships for k , and kb for variation of the leaving phenolate ion are slightly better with u- whereas Hammett u fits K , and the parameters k , and k b for variation in the amido side chain. The values are recorded in Table 111. Trapping with '%labeled water (4.64% la0enrichment) of the hydrolysis of 4-nitrophenyl 2-acetamidobenzenesulfonate, followed by treatment of the solfonate product with 4-toluidine, gave N-acetyl-4-toluidine, which had 4.97% l80labeling. The product from naturally occurring water has 0.664% l80labeling (derived from the small natural lSO abundance). The label was thus incorporated to the extent of [(4.97 - 0.664)/4.64] 100% = 92.8%. The 7.2% difference from 100% may be due to part of the reaction not giving rise to the labeled product or, most likely, to error. Acetanilides, under the same conditions of time and temperature, are not expected to incorporate laO from water in significant amounts. Thermodynamic activation parameters for the hydrolysis of the 4-nitrophenyl benzamidobenzenesulfonateswere
r = 0.999 r = 0.996 r = 0.997
1 M ionic strength made up with KC1, 25 O C . bValues of pK, for 0.2 M ionic strength are included as there does not appear to be a marked salt dependence of pK, here. cThe u- correlation is marginally superior to the u correlation ( r = 0.974 for k, and 0.983 for kb).
Table IV. Thermodynamic Activation Parameters for the Hydrolysis of 4-Nitrophenyl Benzamidobenzenesulfonates -AS* / eu A?I*/ kcal mol-' mol-' substrate ortho ester k, 18.1 f 0.3 4.0 f 1.0 kb 18.3 f 0.4 6.3 f 1.4 kdc,/Kwb 13.9 f 0.1 6.9 f 0.5 para ester knK.¶IKwC 14.1 f 0.7 21.3 f 2.9 OAt 25 O C . "From kom/[OH-] at pH 10.2 (carbonate buffer 0.01 M). cFrom kOM/[OH-]at pH 11.8 (KOH buffer 0.0025 M).
calculated for both the ortho and the para ester and are shown in Table IV. In the para case, the term measured is k&,/K,, since calculations for 25 "C show that the ester is almost completely undissociated in the conditions used, and the contribution from k b to the overall rate is negligible. In the case of the ortho ester, activation parameters were obtained for k,, kb, and k,K,/K,. The results of trapping experiments with ammonia are given in Table V. The hydrolysis of 4-nitrophenyl 2benzamidobenzenesulfonate at pH 10.25 in the absence of ammonia buffer yields a UV spectrum with a peak at 400
3340 J . Org. Chem., VoE. 50, No. 18, 1985 Table V. Ammonia Trapping of the Intermediate in the Hydrolysis of 4-Nitrophenyl 2-Benzamidobenzenesulfonate" [NHd PH 104k0tad A264 4.54 0.60 0.73 0 10.25 4.49 0.49 0.02 10.21 0.73 4.39 0.43 0.79 0.1 10.23 4.49 0.38 0.75 0.2 10.25 4.62 0.34 0.75 0.4 10.25 "5.05 X M ester; 0.008 M carbonate buffer, temperature 25 "C, ionic strength maintained at 1 M with KC1.
nm corresponding to the 4-nitrophenolate ion and a maximum at 264 nm attributable to the 2-benzamidobenzenesulfonic acid anion. As the ammonia concentration is increased there is a decrease in the benzamidobenzenesulfonate product as deduced from the decrease in the absorbance at 264 nm. A t 0.4 M ammonia the product spectrum shows only a slight shoulder in the 264-nm region. It proved impossible for us to prepare the benzamidobenzenesulfonate product by a different route to compare authentic spectra, but benzanilide has an absorption maximum in a similar region with a similar strength (log z 4.2 at 267 nm).13 No change in the rate constant for release of the 4-nitrophenol from t.he 2benzamidobenzenesulfonate ester was observed on using benzylamine buffers up to 0.4 M in 50% aqueous dioxan at pH 10.15 either. These experiments show that the products of trapping vary with the concentration of trapping agent while the rate constant is unaltered. Hydrolysis of the 4-nitrophenyl ester of 2-benzamidobenzenesulfonic acid for 2 days at 70 "C in the presence of 0.1 M KF (phosphate buffer at pH 7) gave no indication of orthanilic acid liberation, as judged by the very sensitive HN02/2-naphthol test." Discussion The kinetic law for the ortho sulfonate ester hydrolysis suggests that two different mechanisms represented by the parameters k , and k b could account for the reaction. We propose that the k, term comes from the mechanism shown in eq 7, where the amido group ionizes (K,) to give an anion
Thea et al. slow under the conditions of the experiment. Acetanilide has a half-life of about 6 h,14 whereas the hydrolysis was carried out in 20 min and had a half-life of less than 28 s. Labeling at the amide oxygen must therefore arise through the intervention of an intermediate such as 2. It would be difficult to explain incorporation either by adventitious labeling while a normal BAc2 mechanism occurs at the sulfur or by formation of a tetrahedral intermediate followed by intramolecular participation as in eq 8. The OH
a
NHCOR
0:
NH'/-'R
on-
s02x
'0-
sopx
3
-x-
aso2,0 -NH$
4
latter mechanism can also be excluded since the putative bimolecular rate constant (k&,/K,) of 1.17 M-' scl for the 4-nitrophenyl 2-acetamidobenzenesulfonate is about 16000-fold faster than the rate constant for addition of M-I s-l at 25 OC).14 hydroxide ion to acetanilide (7.10 X The second-order rate constant (k,K,/K,) for the hydrolysis of 4-nitrophenyl2-acetamidobenzenesulfonateis some 40-fold larger than that of the N-methylated anaM-' s-l (see Table II)]. logue [1.17 as opposed to 2.85 X While this is by no means conclusive evidence, because of the extra steric requirements of the alkylated ester, it is certainly consistent with the proposed scheme (eq 7). Temperature coefficients for the hydrolysis of 4'-nitrophenyl bezamidobenzenesulfonates carried out in the pH region where k , prevails provide strong support to the mechanisms proposed for the two esters.15 Experiments with ammonia as the trapping agent indicate that while the rate constant for release of the phenol is unchanged with increasing amine concentration, the product composition is markedly altered. The results summarized in Table V suggest that the rate-limiting step (phenol release) precedes a subsequent product forming stage. The site of attack of the nucleophile must be at the imino function in 2, otherwise the strong absorbance of the benzanilide would not be destroyed. This site must be + N H ~ h
which expels the leaving group X- in a unimolecular step k, to give the intermediate. Comparison of the reactivity of HO- toward the ortho and para isomers of 4'-nitrophenyl benzamidobenzenesulfonate reveals a ca. 1000-fold ratio between the second-order rate constants. The mechanism for the para ester can only be of the BAc2-type (equivalent to koH in eq 7), and the rate constant for this isomer (6.13 10-3 M-' s-1) comes close to that for the 4-nitrophenyl benzenesulfonate (Table 11). The term kOHin the above mechanism provides only a minor contribution to the reaction flux for the ortho ester. Oxygen-18 labeling studies are consistent with the intervention of an intermediate (2, R = CH3) in the hydrolysis of the 2-acetamidobenzenesulfonate. The amido oxygen in the product from the reaction in 180-enriched water is, within the limits of the experimental error, completely labeled. Oxygen exchange at the amide will be very
that attacked in the trapping by water, otherwise leO would not be incorporated into the amide group. These results are substantially the same as those obtained by Duchek5 on his isolated intermediate. Aberlin and Bunton4 found that the amide group was cleaved in the hydrolysis of the 2-acetamidobenzenesulfonylfluoride. This cannot be due to attack of the fluoride ion (from the early stage of the reaction) on the intermediate (2, R = CH3) at the imino group (eq 10). Duchek5observed that the imino fluoride 5 readily hydrolyzed to the acetanilide (as did the intermediate 2 itself) when exposed to moisture. We find that the hydrolysis of 4-nitrophenyl 2-benzamidobenzene-
(13) Weast, R. C., Ed. "Handbook of Chemistry and Physics", 55th ed.; Chemical Rubber Publishing Co.: Cleveland, OH, 1974-1975; p C-175.
(14)Bender, M. L.; Thomas, R. J. J.Am. Chem. SOC.1961,83,4183. (15) Page, M. I. Chem. SOC.Reu. 1973, 2, 295.
2
J. Org. Chem., Vol. 50, No. 18, 1985
Catalysis of Sulfonate Ester Hydrolysis
-
7
0
so;
5
sulfonate in phosphate buffer at pH 7 in the presence of 0.1 M KF at 70 "C for 2 days gives no orthanilic acid. During the hydrolysis of the o-sulfonyl fluoride4 the medium should rapidly become acidic, owing to the absence of buffer. A reasonable hypothesis about the amido fission observed by Aberlin and Bunton4 is that the fluoride ion attacks the intermediate (2, R = CHJ to give a species which breaks down in acid solution to give the amine 6 (eq 10). Under these conditions, the reaction would not proceed via the imidoyl fluoride 5. Simple attack of HO- on the sulfonate group of the anionic ester is most probably the mechanism giving rise to kb in the hydrolysis of 4-nitrophenyl 4-benzamidobenzenesulfonate. It cannot hold for the hydrolysis of aryl 2-acylamidobenzenesulfonates,though. Indeed, the Hammett p value for variation of the aryl amide group is negative (-1.24) by an amount that cannot be explained simply on the grounds that the equilibrium used to obtain p is not appropriate as a standard model. However, the kb term gives only a minor contribution to the overall reaction flux, even at high KOH concentration. The Hammett p values for variation of the leaving phenol or of the aryl amide group are very much like the corresponding ones derived from k,. Furthermore, the activation entropy is only moderately negative (-6.3 eu mol-') and identical in value (within the limits of the experimental error) with that derived from k,K,/K, (-6.9 eu mol-'). The simplest explanation is that the kb term arises from a specific anion effect, which becomes detectable when a substantial part of the chloride ions is substituted by hydroxide ions. This could be tentatively envisaged as a weak interaction of the type 7 between HOand the ester anion which would give rise to the enhancement of the nucleophilicity of the acylamido anion. Substituent Effects. The substituent effect on pK, is quite marked ( p = -2.02) for variation in the amido function as might have been predicted. It is a little more negative than that found for the ionization of other anilides substituted on the carbonyl function.16 The effect of
I-OH
I
8
3341 I*
9
leaving group substituents on the pK, ( p = -0.46) seems too negative for transmission of the substituent effect through the OSC=CH system. This may be due to an interaction of the type 8. The large negative p value for leaving group variation for k, indicates considerable change in effective charge (-1.28) on the leaving oxygen from ground to transition state. This is similar to that derived from the work of Vizgert for hydroxide ion attack on the aryl esters of benzenesulfonic acid." The change of about -1.3 in a total of -1.8 is consistent with extensive S-O bond fission in the transition state gala The value of p for k, for variation in the amido function suggests that extensive S-0 bond formation could occur in the transition state. The data seem to fit a concerted process at the sulfur atom, involving passage through a transition structure lying on the product side of the reaction.lg Acknowledgment. We are grateful to the S.E.R.C. (England) and C.N.R. (Italy) for financial support. We thank N.A.T.O. for a travel grant (RG 115.80) and the Ciba-Geigy Fellowship Trust for a Senior Fellowship (A.W.). Thanks are due to Dr. L. Tacchi for help with some of the experimental work. Registry No. 1 (R = CH3, X = C6H50),97042-32-5; 1 (R =
CH3, X = 4-NOzC6H,O), 97042-33-6; 1 (R = CsH5, X = 4NOzC6H4O), 97042-36-9; 1 (R = CsH5, X = 4-CNC,H,O), 97042-37-0; 1 (R = CsH5, X = 3-C1C&40), 97042-38-1; 1 (R = C&5, X = 3-BrCsH40), 97042-39-2; 1 (R = CsH5, X = 3CNCBH~O), 97042-40-5; I (R = CsH5, X = C6H50),97042-41-6; 1 (R = 4-CH3CsH4, X = 4-NO&H,), 97042-42-7;1 (R = 3-CICsH4, X = 4-NOzCsH4), 97042-43-8; 1 (R = 3-NOzCsH4, X = 4NO&sH4O), 97042-44-9; ~ - C B H ~ C O N H C ~ H ~ S O ~ O C ~ H ~ - ~ - N O ~ 97042-45-0; C6H5S020C6H4-4-N02, 3313-84-6; 2-nitrobenzenesulfonyl chloride, 1694-92-4; phenyl 2-nitrobenzenesulfonate, 41480-05-1; phenyl 2-aminobenzenesulfonate,68227-69-0; phenyl 2-(N-methylacetamido)benzenesulfonate,97042-34-7; 4-nitrophenyl 2-(N-methylacetamido)benzenesulfonate,97042-35-8; phenol, 108-95-2.
Supplementary Material Available: Analytical and spectroscopic data (Table Sl) and rate data for different temperatures used to calculate activation parameters (Table S2) (3 pages). Ordering information is given on any current masthead page. (16) Williams, A.; Douglas,K. T. J . Chem. SOC.,Perkin Trans. 2 1972, 2112. (17)(a) Thea, S.; Harun, M. G.; Williams, A. J. Chem. SOC.,Chem. Commun. 1979,717.(b) Vizgert, R.V. Zh.Obshch. Khim. 1958,28,1977. (18)(a) Williams, A. In "The Chemistry of Enzyme Action, New Comprehensive Biochemistry"; Elsevier: Amsterdam,1984;Vol6, p 127. (b) Hill, S. V.; Thea, S.; Williams, A. J. Chem. SOC., Perkin Trans. 2 1983, 437. (c) Bourne, N.; Williams, A. J. Org. Chem. 1984,49, 1200. (19)Kice, J. L.Adu. Phys. Org. Chem. 1980,27, 65.
